from espnet2.bin.tts_inference import Text2Speech
from espnet2.utils.types import str_or_none
import argparse, os, time, torch
import scipy.io.wavfile
import shutil

text2speech = Text2Speech.from_pretrained(
    vocoder_tag=str_or_none(vocoder_tag),
    model_file = str_or_none(tag),
    train_config = str_or_none(config),
)

with torch.no_grad():
        start = time.time()
        wav = text2speech(input_text)["wav"]
    rtf = (time.time() - start) / (len(wav) / text2speech.fs)
    print(f"RTF = {rtf:5f}")
    out_file_name = args.out+"/"+str(i)+".wav"
    scipy.io.wavfile.write(out_file_name, text2speech.fs , wav.view(-1).cpu().numpy())